Yu Qina,
Fangxia Shena,
Tianle Zhua,
Wei Honga and
Xiaolong Liu*b
aSchool of Space and Environment, Beihang University, Beijing 100191, China
bBeijing Engineering Research Center of Process Pollution Control, National Engineering Laboratory for Hydrometallurgical Cleaner Production Technology, Institute of Process Engineering, Chinese Academy of Sciences, Beijing 100190, China. E-mail: liuxl@ipe.ac.cn
First published on 27th September 2018
A series of silver catalysts supported on lanthanum based perovskites LaBO3 (B = Co, Mn, Ni, Fe) were synthesized and evaluated in the catalytic oxidation of ethyl acetate. XRD, BET, TEM/HRTEM, HAADF-STEM, XPS and H2-TPR were conducted, and the results indicate that redox activity of the catalysts is of great importance to the oxidation reaction. Activity tests demonstrated that Ag/LaCoO3 was more active than the other samples in ethyl acetate oxidation. Moreover, the CO2 selectivity, COx yields and byproduct distributions for all catalysts were studied, and Ag/LaCoO3 showed the best catalytic performance. Besides, Ag/LaCoO3 also showed excellent catalytic activity for other OVOCs.
Numerous catalysts have been used in the catalytic oxidation of ethyl acetate, and usually include noble metals and metal oxides. Noble metals commonly exhibit high catalytic activity in VOCs abatements.8–10 Au catalysts have been well demonstrated in ethyl acetate oxidation, and Carabineiro et al. reported that Au/oxide catalysts possessed high catalytic activity for the oxidation of ethyl acetate, and pointed out that the role of gold is to enhance the reducibility of the surface of the oxide supports.11 However, the drawbacks of high cost, instability, and sensitivity to poison have greatly limited the practical applications of noble metals in the industrial process. Recently, metal oxides have been intensively studied due to their relatively low cost and poison resistance.12–14 Among the metal oxides, perovskite-type oxides with the general formula of ABO3, have been widely used in air purification,15–18 solar hydrogen production,19 sensor20 and diesel fuel reforming21 due to their advantages of low cost, high thermal stability, and great catalytic performance.22,23
Of all the perovskite-type oxides, lanthanum (La)-based perovskites, namely LaBO3, are recognized as one of the most potential catalysts with higher efficient for their B sites that can be filled with various transition metal cations.24–26 Recently, Taran et al. evaluated the effects of the variation of the B sites in LaBO3 (B = Cu, Fe, Mn, Co, Ni) catalysts for wet peroxide oxidation of phenol, and found that LaCuO3 was the most active one, whereas LaFeO3 was the most stable one.26 Zhu et al. studied a series of A-site substituted La0.8M0.2MnO3 catalysts (M = Ba, Ca, Ce, Mg and Sr) for plasma-catalytic oxidation of ethyl acetate, and found that La0.8Ce0.2MnO3 catalyst has the highest activity, corresponding to its highest reducibility among the tested catalysts.3 Subsequently, Zhu group investigated the plasma-catalytic oxidation of ethyl acetate with La1−xCexCoO3+δ (x = 0, 0.05, 0.1, 0.3 and 0.5) catalysts.1 The removal efficiency of ethyl acetate for the Ce-doped LaCoO3 catalyst was further enhanced, because of the increased specific surface area and a reduced crystallite size compare of pure LaCoO3. Giraudon et al. reported the oxidation of chlorobenzene over Pd/LaBO3 (B = Co, Mn, Fe, Ni), and found that palladium was totally reduced in hydrogen, while the perovskite network was either unaffected or transformed with partial destruction.27 Tang et al. found that LaCoO3 with 3DOM structures exhibited excellent catalytic activity in soot oxidation.28
Meanwhile, silver-based catalysts performed well for the oxidation of various VOCs pollutants, such as ethyl acetate,6 formaldehyde,29 toluene,30 methanol,31 acetone32 and naphthalene.33 Jodaei et al. studied bimetallic Ag–M (M = Fe, Co, Mn)-ZSM-5 for its catalytic oxidation ability of ethyl acetate, and found that Ag–Fe-ZSM-5 has the highest catalytic activity.6 Wang et al. reported that the Ag-doped LaMnO3 perovskite oxide exhibits high activity for simultaneous removal of NOx and diesel because of the abundant oxygen vacancy and over-stoichiometry oxygen in the perovskite lattice along with Ag0.34 However, few work has been done on the use of silver supported LaBO3 perovskite catalysts for ethyl acetate removal. Hence, it is of great interest to investigate the silver-based LaBO3 perovskite catalysts in ethyl acetate oxidation.
In this work, a series of B-site substituted Ag/LaBO3 (B = Co, Mn, Ni, Fe) catalysts were synthesized and evaluated regarding their catalytic oxidation potential of ethyl acetate. The structures of Ag/LaBO3 catalysts were characterized by BET, XRD, TEM/HR-TEM, HAADF-STEM, XPS and H2-TPR, and the role of the B-site substitutes on the reaction performance was further explored. The results indicate that the reducibility of the catalysts plays a key role in the removal of ethyl acetate.
TEM images of Ag/LaCoO3 catalyst are shown in Fig. 2a with the inset of Ag particle size distribution calculated by Image-J software. The Ag particle sizes varied from 2 to 16 nm with the average value at 7.5 ± 2 nm. Moreover, the particle that corresponded to metallic Ag (111) facet with the lattice fringes of 0.24 nm was observed from the high-resolution TEM (HRTEM) (Fig. 2b). The Ag, La and Co distributions were studied by taking HAADF-STEM images and STEM-EDS mapping analysis (Fig. 2c–g). The LaCoO3 nanoparticles were presented as bright spots in Fig. 2c. In the STEM-EDS mappings (Fig. 2d–g), the observed O (purple), Co (yellow), La (green) and Ag (blue) shows similar distributions, indicating that Ag species was well-dispersed on the LaCoO3 support as metallic Ag nanoparticles.
The redox properties of Ag/LaBO3 and LaBO3 catalysts were investigated using H2-TPR from 50 to 600 °C (Fig. 3). The first reduction peaks can be attributed to the reduction of surface oxygen species, which were at 319, 316, 357 and 385 °C for LaCoO3, LaMnO3, LaNiO3 and LaFeO3, respectively (Fig. 3b, d, f and h). In addition, LaCoO3 (1.98 mol g−1) consumes more hydrogen than LaMnO3 (0.45 mol g−1) (Table S2†) because of a large number of adsorption and lattice oxygen on the LaCoO3 surface, suggesting that the sample with LaCoO3 support was easier to be reduced than samples with other supports. For the unsupported LaBO3 perovskites materials, two peaks were observed. The peak at low temperature was ascribed to the reduction of adsorbed oxygen, and the peak at high temperature was due to the reduction of lattice oxygen.3 For Ag/LaCoO3 and LaCoO3, Co3+/Co2+ and Co2+/Co0 transformation occurred in the reduction, and Co3+/Co2+ redox play an important role in the catalytic oxidation, generating abundant mobile oxygen for the realization of the Mars van Krevelen mechanism in ethyl acetate oxidation.11 The low hydrogen consumption by LaFeO3 induced by the low adsorption and lattice oxygen implies that the stable structure of LaFeO3 leads to its lowest oxidizing ability. LaNiO3 has higher hydrogen consumption but lower oxidizing ability than LaMnO3 which is mainly ascribed to the higher first reduction temperature of LaNiO3. Particularly, distinct shifts to low temperatures of the first reduction peak were observed when LaBO3 were doped with silver, indicating better oxygen mobility in silver-based catalysts compared to pure LaBO3. Furthermore, Ag/LaCoO3 catalyst had the highest hydrogen consumption rate of 2.04 mol g−1 (Table S2†), suggesting that the sample possessed the best redox among the tested catalysts.
Fig. 3 H2-TPR profiles of (a) Ag/LaCoO3, (b) LaCoO3; (c) Ag/LaMnO3, (d) LaMnO3; (e) Ag/LaNiO3, (f) LaNiO3; (g) Ag/LaFeO3, (h) LaFeO3. |
XPS spectra were collected to obtain the element chemical states and oxygen vacancies on the catalyst surface. Fig. 4a shows the Ag 3d spectra of the Ag/LaBO3 catalysts. The two peaks with binding energy at 368.3 and 374.3 eV correspond to the Ag 3d5/2 and Ag 3d3/2 states, respectively, which could be attributed to the Ag0 of the metallic Ag. As listed in Table 1, Ag/LaCoO3 showed the highest Ag content (2.64%), the highest ratio of Ag/La (0.14%) and B/La (0.72%) among the catalysts. The surface content of Ag on Ag/LaNiO3 catalyst was 2.50%, which is lower than surface content of Ag/LaCoO3 but higher than those of others, while it possessed the lowest content of B (5.07%). The Ag/La ratio on the surface of Ag/LaMnO3 catalyst was the lowest, while the activity of catalytic oxidation of ethyl acetate was higher than those of Ag/LaNiO3 and Ag/LaFeO3. Unsupported LaMnO3 contributed apparently higher catalytic activity than LaNiO3 and LaFeO3, due to the easy Mn3+/Mn2+ redox, which has been demonstrated in the H2-TPR results. Addition of Ag increased the catalytic activity of the perovskites, revealing that the Ag dispersion and B sites both play important roles in the oxidation process.
Sample | Atomic composition (%) | Ag/La (%) | B/La (%) | Oβ/Ototal (%) | |||
---|---|---|---|---|---|---|---|
Ag | B | La | O | ||||
Ag/LaCoO3 | 2.64 | 13.65 | 18.83 | 64.88 | 0.14 | 0.72 | 41.39 |
Ag/LaMnO3 | 0.74 | 13.26 | 20.57 | 65.43 | 0.04 | 0.64 | 38.12 |
Ag/LaNiO3 | 2.50 | 5.07 | 18.59 | 73.44 | 0.13 | 0.27 | 57.12 |
Ag/LaFeO3 | 1.60 | 13.9 | 20.05 | 64.45 | 0.08 | 0.69 | 17.70 |
The O 1s spectra were measured to obtain the chemical states of oxygen on Ag/LaBO3 catalysts (Fig. 4b). Previous reports showed that the oxygen states in perovskite-like systems were rather complex and ambiguous to interpret.37,38 Nevertheless, the O 1s peak at the lowest binding energy may be assigned to the lattice oxygen O2− and the peak at Eb = 530.8–531.5 eV assigned to oxide defects or surface oxygen ions with low coordination which is denoted as Oβ, while the peak at the highest binding energy is assigned to surface carbonate or water molecules adsorbed on the surface. The relative concentration of Oβ, defined as Oβ/Ototal, was calculated based on the area of the corresponding peaks (Table 1). Ag/LaNiO3 with the highest Oβ/Ototal ratio showed relatively higher surface chemisorbed oxygen, although the catalytic activity was inconsistent with the amount of chemisorbed oxygen. The XPS results here suggest that the chemical states of Ag/LaBO3 catalysts were affected by the B-site of the perovskite, and the catalytic activity was related to the mutual effect of content of Ag0, ratio of Ag/La, B/La and Oβ/Ototal.
In order to evaluate the catalytic stability, on-stream ethyl acetate oxidation for Ag/LaCoO3 was performed at 180, 190 and 200 °C, respectively. The results are summarized in Fig. 6a. When the reaction temperature was 180 °C, the ethyl acetate conversion was stabilized at 58% within 12 h, and no apparent conversion decrease was observed. With the increase of the reaction temperature, the conversion could be maintained at 90 and 99% at 190 and 200 °C, respectively, revealing that Ag/LaCoO3 sample was catalytically durable.
Water vapor commonly exists in the VOCs-containing waste gases, and the anti-moisture ability is crucial to the industrial application of the catalysts. Hence, the anti-moisture experiment was evaluated for Ag/LaBO3 in presence of 1% H2O at 200 °C. As illustrated in Fig. 6b, the conversion was maintained at 99% in the first 4 h. Introduction of 1% H2O led to the decrease of the ethyl acetate conversion from 99% to 88%, while the catalytic activity was recovered when 1% H2O was cut off. Therefore, the water exhibited an inhibitory and reversible effect in the catalytic oxidation of ethyl acetate.
To further investigate the catalytic performance, CO2 selectivity, COx yields, distributions of the organic byproducts for Ag/LaCoO3 and LaCoO3 were studied and compared for Ag/LaCoO3 and LaCoO3, and the results are shown in Fig. 7. Clearly, the addition of Ag significantly promoted the removal of ethyl acetate and the selectivity of CO2, and the CO2 yield are almost equal to the ethyl acetate conversions in the whole experimental temperature for Ag/LaCoO3. Notably, no CO was observed in the catalytic oxidation over Ag/LaCoO3, whereas the CO yield reached 2.8% at 190 °C for LaCoO3. As shown in Fig. 7b and d, acetaldehyde was observed and detected as the main byproduct due to the partial oxidation of ethyl acetate. It is known that ethyl acetate oxidation is a stepwise process with a preliminary hydrolysis to ethanol and acetic acid.1 Further oxidation of ethanol contributed acetaldehyde, which was easily released from the catalyst surface as byproduct. It has been reported that surface acidity facilitates the adsorption of VOCs,39 whereas perovskites materials commonly exhibits weak surface acidity, which might be a reason for the byproduct releasing.40 The maximum concentration of acetaldehyde reached 9.5 and 26.7 ppm at 180 °C for Ag/LaCoO3 and LaCoO3, respectively, suggesting that Ag/LaCoO3 possesses the advantages of higher CO2 selectivity and less byproduct due to Ag addition. For comparison, the CO2 selectivity, COx yields, distributions of the organic byproduct for Ag/LaMnO3 and LaMnO3 (Fig. S1†), Ag/LaNiO3 and LaNiO3 (Fig. S2†), Ag/LaFeO3 and LaFeO3 (Fig. S3†) have been collected and summarized in ESI.† It can be seen that Ag/LaCoO3 and LaCoO3 generates far less byproducts than other perovskite materials at similar conversions, demonstrating their high catalytic activities, which were well-correlated with their physicochemical properties. Addition of Ag commonly decreased the byproduct generations. The maximum concentrations of acetaldehyde were 22.1, 28.1, and 30.8 ppm for Ag/LaMnO3, Ag/LaNiO3, and Ag/LaFeO3, respectively, apparently higher than that of Ag/LaCoO3.
Other kinds of OVOCs such as the ethers, alcohols, and ketones, commonly exist in the industrial waste gases. Hence, the most efficient catalyst Ag/LaCoO3 was also employed in the catalytic oxidation of diethyl ether, ethanol and acetone to test its oxidation ability to various OVOCs abatements. As summarized in Fig. 8, the T90 of ethanol, ethyl acetate, acetone, and diethyl ether were 159 °C, 190 °C, 209 °C, and 239 °C, respectively, demonstrating that Ag/LaCoO3 exhibits outstanding applicability to various OVOCs purifications.
Footnote |
† Electronic supplementary information (ESI) available: Specific surface areas, H2 consumption of the samples, and byproduct distributions. See DOI: 10.1039/c8ra06933f |
This journal is © The Royal Society of Chemistry 2018 |